Gary Marcus

Rank 22 of 47
|
Score 46
Gary Marcus
@GaryMarcus
·
21d
brutal thread: even the safest AI agent got beat 1.5% of the time.

none of this stuff is remotely secure.
7/30/2025, 12:27:55 AM
X
In reply to:
Gary Marcus
@GaryMarcus
·
21d
@tombombadeel yes of course but (exactly as I predicted) they are neither reliable nor secure, and are causing lots of havoc
Tom
@tombombadeel
·
21d
@GaryMarcus there factually are a couple AI agents this year, Gary
Gary Marcus
@GaryMarcus
·
21d
does anybody still think that 2025 is going to be the year of AI agents or we just going to forget people ran around saying that last year?

The conversation involves a discussion on the reliability and security of AI agents, referencing a test where AI agents were breached multiple times. The tone is critical and skeptical, focusing on the potential risks and failures of AI technology.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The statement highlights potential harm caused by AI agents, aligning with the principle of striving to do no harm by raising awareness of security issues. [+1]
  2. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    The discussion promotes understanding of AI security challenges, fostering awareness and caution. [+1]
  3. Principle 4:
    I will engage in constructive criticism and dialogue with those in disagreement and will not engage in personal attacks or ad hominem arguments.
    The conversation engages in constructive criticism of AI technology, focusing on its shortcomings without personal attacks. [+1]